Increased reflex bronchoconstriction is a characteristic of asthma. We have recently shown a doubling of airway epithelial sensory nerves in a mouse model of asthma. This increased reflex bronchoconstriction substantially. Eliminating eosinophils prevented the increases in both innervation and bronchoconstriction. We also showed that eosinophils increase the growth of cultured dorsal root ganglion sensory neurons, due to a soluble factor produced by the eosinophils. Thus eosinophils increase airway sensory innervation, and this may participate in both the bronchoconstriction and the cough that are common in asthma. We hypothesize that eosinophils promote airway sensory nerve growth and this increases reflex bronchoconstriction and cough. We propose three specific aims: SPECIFIC AIM #1: To define eosinophil-mediated effects on airway sensory innervation. In both an antigen challenge model and unique transgenic models of eosinophilic pulmonary inflammation, we will use our novel whole-mount nerve imaging [8] to define eosinophil-dependent changes in airway epithelial sensory innervation, as well as on substance P content, and TRPA1, TRPV1, TRPM8, and ASIC3 expression will be measured. These will be correlated with airway reflex bronchoconstriction. SPECIFIC AIM #2: To determine the reversibility of these pulmonary remodeling events by targeting eosinophils in mice with established disease. We will use our new strains of mice that allow inducible on demand ablation of eosinophils (the iPHIL mouse). We have also generated mice with eosinophils that lack the glucocorticoid receptor, allowing us to determine whether the effects of steroid treatment are mediated by suppressing eosinophils or via eosinophil independent pathways. SPECIFIC AIM #3: To characterize neural plasticity in human airway disease. Similar histological endpoints as in aims 1 and 2 will be measured in biopsies from well-characterized patients with 1) asthma (mild and severe), 2) cough-variant asthma, and 3) cough without asthma, comparing these with normal volunteers. Induced sputum and bronchoalveolar lavage eosinophil peroxidase will be measured as possible biomarkers.